Chromatographic quantitative rapid analysis method, system, and medium

Abstract

This invention discloses a rapid quantitative chromatographic analysis method, system, and medium, belonging to the field of in vitro rapid diagnostic technology. The method includes: converting the immunochromatographic image of the sample to be tested into a grayscale image, and searching for C- and T-line ROI regions in the grayscale image; compressing the first region either as a whole or in sections along the vertical direction of the fluid flow to obtain a corresponding one-dimensional curve, wherein the first region is any one of the C- or T-line ROI regions; dividing the one-dimensional curve into two sub-curves using the highest point of the one-dimensional curve as the boundary, and calculating the inflection point of each sub-curve; generating two boundary lines based on the inflection points obtained in the first region, and using the area between the two boundary lines as a new first region to obtain a new C-line ROI region and a new T-line ROI region; analyzing the concentration of the sample to be tested based on the grayscale values ​​of the new C-line ROI region and the new T-line ROI region. This method can obtain sub-pixel-level signal extraction results.

Description

Technical Field

[0001] This invention belongs to the field of in vitro rapid diagnostic technology, and more specifically, relates to a chromatographic quantitative rapid analysis method, system, and medium. Background Technology

[0002] Immunochromatography achieves detection by triggering an antigen-antibody binding reaction during chromatography. Due to its advantages such as simplicity, speed, no need for personnel training, and minimal or no equipment requirements, immunochromatography has been widely applied in various fields including food testing, drug detection, environmental monitoring, and clinical diagnosis.

[0003] By capturing and processing images of test strips using optical devices, automated qualitative or quantitative analysis of immunochromatographic test strip results can be achieved. Specifically, the test strip is illuminated by a light source, and a photosensitive element collects the reflected or transmitted light signals from the test strip to obtain a colorimetric image. Subsequent image processing methods then analyze the colorimetric image to obtain the judgment result.

[0004] Existing image processing methods are all based on pixel-level quantitative analysis, which yields good feature extraction results for high-resolution images. However, for low-resolution images, due to problems such as high noise and blurred edges, the performance of existing methods is often unsatisfactory. Therefore, there is an urgent need for a high-precision quantitative image analysis method. Summary of the Invention

[0005] In response to the shortcomings and improvement needs of existing technologies, this invention provides a rapid quantitative analysis method, system, and medium for chromatography, aiming to solve the problem that existing methods are difficult to accurately extract signals when processing low-resolution images.

[0006] To achieve the above objectives, according to one aspect of the present invention, a rapid quantitative chromatographic analysis method is provided, comprising: converting an immunochromatographic image of a sample to be tested into a grayscale image, and searching for a C-line ROI region and a T-line ROI region in the grayscale image; compressing a first region either as a whole or in sections along the direction perpendicular to the liquid flow direction to obtain a corresponding one-dimensional curve, wherein the first region is either the C-line ROI region or the T-line ROI region; dividing the one-dimensional curve into two sub-curves using the highest point of the one-dimensional curve as a boundary, and calculating the inflection point of each sub-curve; generating two boundary lines based on the inflection points obtained in the first region, and using the region between the two boundary lines as a new first region to obtain a new C-line ROI region and a new T-line ROI region; and analyzing the concentration of the sample to be tested based on the grayscale values ​​of the new C-line ROI region and the new T-line ROI region.

[0007] Furthermore, when the first region is compressed as a whole: there is a one-dimensional curve corresponding to the first region; there are two inflection points obtained in the first region; and the two generated boundary lines are straight lines that pass through the inflection points and are perpendicular to the direction of liquid flow.

[0008] Furthermore, when the first region is compressed by region: there are N one-dimensional curves corresponding to the first region, N≥2; there are 2N inflection points obtained in the first region; generating two boundary lines based on the inflection points obtained in the first region includes: smoothly connecting the inflection points located on the same side of the highest point of the one-dimensional curve in sequence, and using the two curves obtained by the connection as two boundary lines.

[0009] Furthermore, the regional compression of the first region includes: dividing the first region vertically into groups every k columns of pixels along the vertical direction of the liquid flow direction, so as to divide the first region into N sub-regions, 1≤k≤7; and performing overall compression on each sub-region along the vertical direction of the liquid flow direction to obtain a one-dimensional curve corresponding to each sub-region.

[0010] Furthermore, the step of searching for the C-line ROI region and the T-line ROI region in the grayscale image includes: sequentially performing compression processing and moving mean filtering processing on the grayscale image in the direction perpendicular to the liquid flow direction to obtain the corresponding one-dimensional smooth curve; searching for the peak point of the one-dimensional smooth curve, wherein the value of the peak point satisfies:

[0011] S peak >0.2*max(S)+0.8*min(S)

[0012] Among them, S peak Let S be the value of the peak point, and S be the one-dimensional smooth curve. Find the peak endpoints along both sides of the one-dimensional smooth curve from the peak point, and determine the C-line ROI region and the T-line ROI region based on the peak endpoints and the peak point.

[0013] Furthermore, the step of analyzing the concentration of the sample to be tested based on the gray values ​​of the new C-line ROI region and the new T-line ROI region includes: calculating the C-line signal value based on the gray values ​​of the new C-line ROI region and the C-line background region; calculating the T-line signal value based on the gray values ​​of the new T-line ROI region and the T-line background region; and calculating the concentration of the sample to be tested based on the ratio between the T-line signal value and the C-line signal value.

[0014] Furthermore, the C-line signal value and the T-line signal value are respectively:

[0015] C = |g b -g0|

[0016] T=|g′ b -g′0|

[0017] Where C is the C-line signal value, T is the T-line signal value, and g b g' represents the grayscale value of the new C-line ROI region, g0 represents the grayscale value of the C-line background region, and g' represents the grayscale value of the background region. b is the grayscale value of the new T-line ROI region, and g′0 is the grayscale value of the T-line background region.

[0018] Furthermore, the C-line signal value and the T-line signal value are respectively:

[0019] C=|lg(g b )-lg(g0)|

[0020] T=|lg(g′ b )-lg(g′0)|

[0021] Where C is the C-line signal value, T is the T-line signal value, and g b g' represents the grayscale value of the new C-line ROI region, g0 represents the grayscale value of the C-line background region, and g' represents the grayscale value of the background region. b is the grayscale value of the new T-line ROI region, and g′0 is the grayscale value of the T-line background region.

[0022] According to another aspect of the present invention, a rapid chromatographic quantitative analysis system is provided, comprising: a conversion and search module for converting an immunochromatographic image of a sample to be tested into a grayscale image, and searching for a C-line ROI region and a T-line ROI region in the grayscale image; a compression module for compressing a first region either as a whole or in sections along the direction perpendicular to the liquid flow direction to obtain a corresponding one-dimensional curve, wherein the first region is either the C-line ROI region or the T-line ROI region; an inflection point calculation module for dividing the one-dimensional curve into two sub-curves with the highest point of the one-dimensional curve as the boundary, and calculating the inflection point of each sub-curve; a generation and update module for generating two boundary lines based on the inflection points obtained in the first region, and using the region between the two boundary lines as a new first region to obtain a new C-line ROI region and a new T-line ROI region; and an analysis module for analyzing the concentration of the sample to be tested based on the grayscale values ​​of the new C-line ROI region and the new T-line ROI region.

[0023] According to another aspect of the invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the rapid quantitative chromatographic analysis method as described above.

[0024] In summary, the above-described technical solutions conceived in this invention can achieve the following beneficial effects:

[0025] (1) A rapid quantitative analysis method for chromatography is provided. After obtaining the grayscale image of the immunochromatographic image, the approximate location of the detection line (C line) and the control line (T line) (i.e. their respective location areas) is quickly located. Then, the grayscale value curve of the location area is used to determine the boundary point, thereby determining the precise boundary line of the Region of Interest (ROI). Finally, a more accurate background area and region of interest are obtained to improve the analytical accuracy of the concentration of the sample to be tested. It can still ensure high analytical accuracy when the resolution of the immunochromatographic image is low.

[0026] (2) The method proposes to compress the ROI region by region, use the gray value curve of each sub-location region to determine more boundary points, and smoothly connect multiple boundary points to obtain more accurate boundary lines, which further improves the positioning accuracy of the background region and the region of interest, thereby further improving the analysis accuracy.

[0027] (3) In a normal detection reaction, there is only one place where the slope of the curve corresponding to the C line area is 0. If the slope of the curve is always 0, the result of abnormal detection reaction can be directly output, which makes it easier for staff to troubleshoot the fault in the detection reaction in a timely manner and improve the overall detection efficiency.

[0028] (4) Compared with quantitative analysis of immunochromatographic test strips through machine learning, on the one hand, the amount of computation is reduced and the detection rate is increased, and on the other hand, the region of interest is accurately located, which improves the accuracy of the detection results; and it is not limited by the number of T lines, the acquisition and segmentation of each T line image is easy, and it can be used to process multi-test cards. Attached Figure Description

[0029] Figure 1 A flowchart of a rapid quantitative analysis method for chromatography provided in an embodiment of the present invention;

[0030] Figure 2A , Figure 2B , Figure 2C , Figure 2D , Figure 2E These are schematic diagrams of the immunochromatographic image, grayscale image, initial ROI region, new ROI region, and concentration-TC relationship curve in Embodiment 1 of the present invention.

[0031] Figure 3 This is a schematic diagram of the one-dimensional curve, inflection point, and boundary line obtained in the process of calculating the new ROI region based on overall compression in Embodiment 1 of the present invention.

[0032] Figure 4A , Figure 4B , Figure 4C , Figure 4D , Figure 4EThese are schematic diagrams of the immunochromatographic image, grayscale image, initial ROI region, new ROI region, and concentration-TC relationship curve in Embodiment 2 of the present invention.

[0033] Figure 5 This is a schematic diagram of the one-dimensional curve, inflection point, and boundary line obtained in the process of calculating a new ROI region based on regional compression in Embodiment 2 of the present invention.

[0034] Figure 6A , Figure 6B , Figure 6C , Figure 6D , Figure 6E These are schematic diagrams of the immunochromatographic image, grayscale image, initial ROI region, new ROI region, and concentration-TC relationship curve in Embodiment 3 of the present invention.

[0035] Figure 7 This is a schematic diagram of the one-dimensional curve, inflection point, and boundary line obtained in the process of calculating the new ROI region based on overall compression in Embodiment 3 of the present invention.

[0036] Figure 8A , Figure 8B , Figure 8C , Figure 8D , Figure 8E These are schematic diagrams of the immunochromatographic image, grayscale image, initial ROI region, new ROI region, and concentration-TC relationship curve in Embodiment 4 of the present invention.

[0037] Figure 9 This is a schematic diagram of the one-dimensional curve, inflection point, and boundary line obtained in the process of calculating the new ROI region based on regional compression in Embodiment 4 of the present invention.

[0038] Figure 10 This is a block diagram of a rapid quantitative chromatographic analysis system provided in an embodiment of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0040] In this invention, the terms "first," "second," etc. (if present) in the invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0041] Figure 1This is a flowchart of the chromatographic quantitative rapid analysis method provided by an embodiment of the present invention. Refer to Figure 1 and combine with Figures 2A-9 to elaborate on the chromatographic quantitative rapid analysis method in this embodiment in detail. The method includes operations S1 - S5.

[0042] Operation S1: Convert the immunochromatographic image of the test sample into a grayscale image, and search for the C-line ROI area and the T-line ROI area in the grayscale image.

[0043] Drop the test sample onto the sample well of the immunochromatographic test strip card. After waiting for several minutes (usually from a few minutes to dozens of minutes, depending on the reaction stabilization time of the test strip card), capture the image within the detection window through an image acquisition device to obtain the immunochromatographic image of the test sample. The pixel size of the image is m×n, indicating that the image data has m rows and n columns. You can first photograph the detection kit, then obtain the detection window image, and then obtain the positioning area image.

[0044] The immunochromatographic test strip card is, for example, an immunochromatographic test strip card that can be显色 under natural light such as colloidal gold, colloidal carbon, colloidal selenium, etc., or a fluorescence immunochromatographic test strip card that requires excitation by light of a specific wavelength. The image acquisition device is, for example, a common image capture device such as a CCD camera, a CMOS camera, a CIS camera, etc.

[0045] Convert the immunochromatographic image into a grayscale image through the grayscale conversion formula Gray = A*r + B*g + C*b, where r, g, and b represent the three channels of the RGB color space, and A, B, and C represent the proportions occupied by the three channels of the RGB color space.

[0046] In the grayscale conversion formula, A + B + C = 1. In different situations, the values of A, B, and C are different. For example, in the image of a colloidal gold immunochromatographic test strip card, usually C < A < B; in the image of a fluorescence immunochromatographic test strip card, the values of A, B, and C are determined by the color characterized by the signal. Whichever color the signal characterizes, the proportion of that color channel is higher.

[0047] In this embodiment, use peak detection to search for the C-line ROI area and the T-line ROI area in the grayscale image, specifically including sub-operations S11 - S13.

[0048] In sub-operation S11, perform vertical compression processing along the liquid flow direction and moving average filtering processing on the grayscale image in sequence to obtain the corresponding one-dimensional smooth curve.

[0049] For the grayscale image of an immunochromatographic test strip card that can be显色 under natural light, the one-dimensional curve formula obtained by performing vertical compression processing along the liquid flow direction on the grayscale image is:

[0050] 需注意,原文中“显色”一词未明确具体含义,这里直接保留了中文表述。你可根据实际情况进行调整。

[0051] For the image of the fluorescence immunochromatographic test strip, the grayscale image is compressed in the direction perpendicular to the liquid flow direction, and the resulting one-dimensional curve formula is:

[0052]

[0053] The formula for moving average filtering is as follows:

[0054] S i =(C i-3 +…+C i +…+C i+3 ) / 7

[0055] Among them, C i Let G represent the value at the i-th point on a one-dimensional curve. i，j S represents the grayscale value of the pixel in the i-th row and j-th column of a grayscale image. i Let S represent the value at the i-th point on a one-dimensional smooth curve. The value is obtained by moving mean filtering. i It can also be for other cases, such as taking (C) i-2 +...+C i +...+C i+2 ) / 5 etc.

[0056] In suboperation S12, the peak point of the one-dimensional smooth curve is searched, and the value of the peak point satisfies:

[0057] S peak >0.2*max(S)+0.8*min(S)

[0058] Among them, S peak Let S be the value at the peak point, and let S be a one-dimensional smooth curve. max(S) and min(S) are the maximum and minimum values ​​of points on the one-dimensional smooth curve, respectively.

[0059] In this embodiment, for example according to S i-3 <S i-2 <S i-1 <S i >S i+1 >S i+2 >S i+3 There are other ways to find peak points.

[0060] In sub-operation S13, the peak endpoints are found along both sides of the one-dimensional smooth curve from the peak point, and the C-line ROI region and the T-line ROI region are determined based on the peak endpoints and the peak point.

[0061] In this embodiment, the method for finding the peak endpoint is, for example, that if the value of a certain point is less than 40% of the value of the peak point, and twice the value of that point is greater than or equal to the sum of the values ​​of two adjacent points, then that point is the peak endpoint. Other methods can also be used to determine the peak endpoint.

[0062] In this embodiment, the C-line ROI region and the T-line ROI region can be obtained separately; alternatively, a single ROI region containing both the C-line and the T-line can be obtained, followed by an additional step of separating the two ROI regions.

[0063] Operation S2: Compress the first region either as a whole or in sections along the direction perpendicular to the flow direction to obtain the corresponding one-dimensional curve. The first region is either the C-line ROI region or the T-line ROI region.

[0064] Operation S3 divides the one-dimensional curve into two sub-curves, using the highest point of the one-dimensional curve as the boundary, and calculates the inflection point of each sub-curve.

[0065] Operation S4 generates two boundary lines based on the inflection points obtained in the first region, and takes the region between the two boundary lines as the new first region, thus obtaining the new C-line ROI region and the new T-line ROI region.

[0066] In one embodiment of the present invention, the first region is compressed as a whole, resulting in a one-dimensional curve corresponding to the first region; two inflection points are obtained in the first region; and two boundary lines are generated, which are straight lines passing through the inflection points and perpendicular to the liquid flow direction. The region between the two boundary lines is taken as the accurate boundary of the first region.

[0067] Taking the C-line ROI region as an example, the specific implementation process of operations S2-S4 under the overall compression method is explained. The C-line ROI region is compressed along the direction perpendicular to the fluid flow direction to obtain the corresponding one-dimensional curve. The principle is the same as the compression processing principle in sub-operation S11. The resulting one-dimensional curve is shown below. Figure 3 The curve is located on the left side. Using the highest point of the one-dimensional curve as the boundary, the curve is divided into two sub-curves. Least squares fitting is performed on each sub-curve, and the inflection point of the fitted curve is calculated. This inflection point is the boundary point (edge ​​point). The obtained inflection points are as follows: Figure 3 The curve is located in the middle. Along the inflection point, draw a line perpendicular to the direction of fluid flow; this is the boundary line. Two inflection points correspond to two boundary lines, as shown below. Figure 3 The area between the two boundary lines in the right-hand figure is the final C-line ROI region.

[0068] In another embodiment of the present invention, a first region is compressed by region, and there are N one-dimensional curves corresponding to the first region, where N≥2; the number of inflection points obtained in the first region is 2N. Generating two boundary lines based on the inflection points obtained in the first region includes: sequentially and smoothly connecting the inflection points located on the same side of the highest point of the one-dimensional curve, and using the two connected curves as two boundary lines. The region between the two boundary lines is taken as the accurate boundary of the first region.

[0069] The first region compression involves: vertically dividing the first region into N sub-regions along the vertical direction of the liquid flow, with every k columns of pixels forming a group, where 1 ≤ k ≤ 7; then, compressing each sub-region as a whole along the vertical direction of the liquid flow to obtain a one-dimensional curve for each sub-region. If the number of remaining pixel columns is less than k, it is also grouped together, resulting in a total of [n / k] sub-regions.

[0070] Taking the C-line ROI region as an example, the specific implementation process of operations S2-S4 under the regional compression method is explained. Vertically along the direction of the fluid flow, the C-line ROI region is divided into groups every k columns of pixels, thus dividing the C-line ROI region into N sub-regions, where 1 ≤ k ≤ 7, and N = [n / k], where [n / k] represents the smallest integer greater than or equal to n / k. Vertically along the direction of the fluid flow, each sub-region is compressed as a whole, resulting in N sets of one-dimensional curves. The principle is the same as the compression processing principle in sub-operation S11. The resulting N sets of one-dimensional curves are shown below. Figure 5 The curve located on the left is considered. For each one-dimensional curve, it is divided into two segments with its highest point as the boundary. Least squares fitting is performed on each segment, and the inflection point of the fitted curve is calculated. This inflection point is the boundary point (edge ​​point). There are 2N inflection points for N sets of one-dimensional curves. Figure 5 The curve is located in the middle. Preferably, all boundary points are marked, and severely deviated boundary points are removed. The remaining adjacent boundary points on the same side as the highest point are connected to form two boundary lines, such as... Figure 5 The area between the two boundary lines in the right-hand figure is the final C-line ROI region.

[0071] Preferably, in this embodiment, the sub-curve is fitted using the least squares method with an S-curve fitting formula. Examples of S-curves include the Logistics curve, Gompertz curve, Weibull curve, Richard curve, Morgan-Mercer-Flodin curve, and Hill curve. The Logistics curve formula is:

[0072]

[0073] The inflection point of the Logistics curve is:

[0074]

[0075] The formula for the Gompertz curve is:

[0076]

[0077] The inflection point of the Gompertz curve is:

[0078]

[0079] The formula for the Weibull curve is:

[0080]

[0081] The inflection point of the Weibull curve is:

[0082]

[0083] The formula for the Richard curve is:

[0084]

[0085] The inflection point of the Richard curve is:

[0086]

[0087] The formula for the Morgan-Mercer Flodin curve is:

[0088]

[0089] The inflection point of the Morgan-Mercer Flodin curve is:

[0090]

[0091] The formula for the Hill curve is:

[0092]

[0093] The inflection point of the Hill curve is:

[0094]

[0095] Operate S5 to analyze the concentration of the sample to be tested based on the gray values ​​of the new C-line ROI region and the new T-line ROI region.

[0096] According to an embodiment of the present invention, operation S5 includes sub-operations S51-S53.

[0097] In sub-operation S51, the C-line signal value is calculated based on the grayscale value of the new C-line ROI region and the grayscale value of the C-line background region.

[0098] In sub-operation S52, the T-line signal value is calculated based on the grayscale value of the new T-line ROI region and the grayscale value of the T-line background region.

[0099] In one embodiment of the present invention, the C-line signal value and the T-line signal value are respectively:

[0100] C = |g b -g0|

[0101] T = |g ′ b -g ′ 0|

[0102] Where C is the C-line signal value, T is the T-line signal value, and g b g is the grayscale value of the new C-line ROI region, g0 is the grayscale value of the C-line background region, g ′ b For the grayscale value of the new T-line ROI region, g ′ 0 represents the grayscale value of the T-line background area. For fluorescent immunochromatographic test strips, the C-line and T-line signal values ​​are calculated in this way.

[0103] In another embodiment of the present invention, the C-line signal value and the T-line signal value are respectively:

[0104] C=|lg(g b )-lg(g0)|

[0105] T=|lg(g ′ b )-lg(g ′ 0)|

[0106] For colorimetric immunochromatographic test strips that can be developed under natural light, either of these two methods can be chosen to calculate the C-line signal value and the T-line signal value.

[0107] In sub-operation S53, the concentration of the sample to be tested is calculated based on the ratio between the T-line signal value and the C-line signal value.

[0108] Calculate the TC value based on the T-line signal value and the C-line signal value:

[0109] TC = T / C

[0110] Then, based on the standard curve, the concentration of the test sample is calculated using the T / C value. The standard curve can be obtained as follows: First, use h immunochromatographic test strips identical to the test strips for the analyte to test h different solutions of known concentrations, where h is a natural number between 4 and 10; then, obtain the TC value in the same way as with the test strips; finally, use a four-parameter Logistic formula for curve fitting to obtain the standard curve of TC value versus analyte concentration. The four-parameter Logistic formula is:

[0111]

[0112] The calibration method for the concentration of the analyte on the immunochromatographic test strip is as follows: find the concentration value corresponding to the point in the TC value-concentration standard curve of the analyte that is the same as the TC value of the immunochromatographic test strip. This concentration value is the concentration of the sample to be tested.

[0113] Example 1

[0114] The methamphetamine sample is dropped onto the sample well of the methamphetamine fluorescent immunochromatographic test strip. After waiting for 5 minutes, the image within the detection window is captured by a CCD camera (image size 540×110 pixels). Figure 2A As shown. The original image is converted to a grayscale image using the grayscale conversion formula Gray = 0.9*r + 0.05*g + 0.05*b, as shown. Figure 2B As shown.

[0115] Peak detection is used to search for C-line and T-line ROI regions in a grayscale image, such as... Figure 2C As shown. The accurate boundary of the signal is calculated using a curve fitting method, as shown... Figure 2D As shown.

[0116] The specific process of the curve fitting method is as follows: First, compress the ROI region (either the C-line ROI region or the T-line ROI region) along the direction perpendicular to the fluid flow direction, that is, calculate the mean of each row to obtain the corresponding one-dimensional curve, such as... Figure 3 The curve is located on the left side; then, the one-dimensional curve is divided into two segments from the highest point. The least squares method is used to fit each segment using the Logistic curve formula, and the inflection point of each segment is calculated based on the fitting results. Figure 3 The curve is located in the middle; finally, along the boundary point, draw a line perpendicular to the direction of liquid flow, which is the boundary line, as shown below. Figure 3 The image is located on the right side of the middle section.

[0117] Using C=|g b -g0|、T=|g ′ b -g ′0| Calculate the C-line signal value and T-line signal value C, T, C = 78.83, T = 100.73. TC = T / C = 1.27. Then, based on the standard curve, use the TC value to calculate the concentration of methamphetamine in the test sample as 4.56 ng / mL.

[0118] Method for obtaining the standard curve: Six methamphetamine solutions of known concentrations were tested using six methamphetamine fluorescent immunochromatographic test strips. The methamphetamine solution concentrations were 0, 1, 2.5, 7.5, 15, and 25 ng / mL, respectively. The characteristic value TC was obtained using the same method as for the immunochromatographic test strips. A standard curve of TC versus methamphetamine concentration was obtained by fitting a four-parameter logistic curve. Figure 2E As shown.

[0119] Example 2

[0120] The methamphetamine sample is dropped onto the sample well of the methamphetamine fluorescent immunochromatographic test strip. After waiting for 5 minutes, the image within the detection window is captured using a CMOS camera (image size 540×110 pixels). Figure 4A As shown. The original image is converted to a grayscale image using the grayscale conversion formula Gray = 0.9*r + 0.05*g + 0.05*b, as shown. Figure 4B As shown.

[0121] Peak detection is used to search for C-line and T-line ROI regions in a grayscale image, such as... Figure 4C As shown. The accurate boundary of the signal is calculated using a curve fitting method, as shown... Figure 4D As shown.

[0122] The specific process of the curve fitting method is as follows: First, the ROI region is divided into groups of 5 columns of pixels along the direction perpendicular to the flow direction, resulting in a total of N = 22 groups of regions; then, each group of regions is compressed as a whole along the direction perpendicular to the flow direction to obtain N sets of one-dimensional curves, such as... Figure 5 The curve located on the left is then divided into two segments from the highest point of each of the N one-dimensional curves. The Richards curve fitting formula is then used to fit each segment using the least squares method. The inflection points of each segment are calculated based on the fitting results. Figure 5 The curve is located in the middle; finally, all boundary points are marked on the grayscale image, and adjacent boundary points are connected to form boundary lines, such as... Figure 5 The image is located on the right side of the middle section.

[0123] Using C=|g b -g0|、T=|g ′ b -g ′0|, calculated to be C=82.07, T=62.48, TC=0.76; based on the standard curve, the concentration of methamphetamine in the test sample was calculated to be 14.13 ng / mL.

[0124] Method for obtaining the standard curve: Six methamphetamine solutions of known concentrations were tested using six methamphetamine fluorescent immunochromatographic test strips. The methamphetamine solution concentrations were 0, 1, 2.5, 5, 7.5, and 25 ng / mL, respectively. The characteristic value TC was obtained using the same method as for the immunochromatographic test strips. A standard curve of TC versus methamphetamine concentration was obtained by fitting a four-parameter logistic curve. Figure 4E As shown.

[0125] Example 3

[0126] The vomitoxin sample was dropped onto the sample well of the vomitoxin colloidal gold immunochromatographic test strip. After waiting for 6 minutes, the image within the detection window was captured using a CMOS camera (image size 370×80 pixels). Figure 6A As shown. The original image is converted to a grayscale image using the grayscale conversion formula Gray = 0.299*r + 0.587*g + 0.114*b, as shown. Figure 6B As shown.

[0127] Peak detection is used to search for C-line and T-line ROI regions in a grayscale image, such as... Figure 6C As shown. The accurate boundary of the signal is calculated using a curve fitting method, as shown... Figure 6D As shown.

[0128] The specific process of the curve fitting method is as follows: Compress the ROI region (either the C-line ROI region or the T-line ROI region) along the direction perpendicular to the fluid flow direction, that is, calculate the mean of each row to obtain the corresponding one-dimensional curve, such as... Figure 7 The curve is located on the left side; the one-dimensional curve is divided into two segments from its highest point. The Gompertz curve formula is used to fit each segment using the least squares method, and the inflection point of each segment is calculated based on the fitting results. Figure 7 The curve located in the middle; along the boundary point, draw a line perpendicular to the direction of liquid flow, which is the boundary line, such as... Figure 7 The image is located on the right side of the middle section.

[0129] Using C=|g b -g0|、T=|g ′ b -g ′ 0|, calculated to obtain C=25.63, T=55.95, TC=2.18, and then according to the standard curve, the concentration of vomitoxin in the test sample was calculated to be 5.10 ng / mL.

[0130] Method for obtaining the standard curve: Seven known concentrations of vomitoxin solutions were tested using seven colloidal gold immunochromatographic test strips for vomitoxin, with concentrations of 0, 2.5, 5, 10, 20, 40, and 80 ng / mL. Characteristic values ​​(TC) were obtained using the same method as for the immunochromatographic test strips. Finally, a standard curve was obtained by fitting a four-parameter logistic curve to the TC values ​​versus the vomitoxin concentration. Figure 6E As shown.

[0131] Example 4

[0132] The vomitoxin sample was dropped onto the sample well of the vomitoxin colloidal gold immunochromatographic test strip. After waiting for 6 minutes, the image within the detection window was captured using a CIS camera (image size 370×80 pixels). Figure 8A As shown. The original image is converted to a grayscale image using the grayscale conversion formula Gray = 0.299*r + 0.587*g + 0.114*b, as shown. Figure 8B As shown.

[0133] Peak detection is used to search for C-line and T-line ROI regions in a grayscale image, such as... Figure 8C As shown. The accurate boundary of the signal is calculated using a curve fitting method, as shown... Figure 8D As shown.

[0134] The specific process of the curve fitting method is as follows: The ROI region is divided into groups of 4 columns of pixels along the direction perpendicular to the fluid flow direction, resulting in a total of N = 20 groups; each group is then compressed along the direction perpendicular to the fluid flow direction to obtain N sets of one-dimensional curves, such as... Figure 9 The curve located on the left is used as an example. Each of the N groups of one-dimensional curves is divided into two segments from its highest point. Then, the Hill curve fitting formula is used to fit each segment using the least squares method. Based on the fitting results, the inflection points of each segment are calculated. Figure 9 The curve located in the middle; mark all boundary points on the grayscale image and connect adjacent boundary points to form boundary lines, such as... Figure 9 The image is located on the right side of the middle section.

[0135] Using C=|lg(g) b )-lg(g0)|、T=|lg(g ′ b )-lg(g ′ 0)|, we calculated C=0.052, T=0.083, TC=1.60, and then calculated the concentration of vomitoxin in the test sample to be 9.61ng / mL according to the standard curve.

[0136] Method for obtaining the standard curve: Seven known concentrations of vomitoxin solutions were tested using seven colloidal gold immunochromatographic test strips for vomitoxin, with concentrations of 0, 1, 2.5, 5, 20, 40, and 80 ng / mL. Total toxicity (TC) was obtained using the same method as with the test strips. A standard curve was obtained by fitting a four-parameter logistic curve to the TC values ​​versus the vomitoxin concentrations, as shown below. Figure 8E .

[0137] Figure 10 This is a block diagram of a rapid chromatographic quantitative analysis system provided in an embodiment of the present invention. (See also...) Figure 10 The rapid quantitative analysis system 1000 includes a transformation and search module 1010, a compression module 1020, an inflection point calculation module 1030, a generation and update module 1040, and an analysis module 1050.

[0138] The conversion and search module 1010, for example, performs operation S1 to convert the immunochromatographic image of the sample to be tested into a grayscale image and search for the C-line ROI region and the T-line ROI region in the grayscale image.

[0139] For example, the compression module 1020 performs operation S2 to compress the first region as a whole or in sections along the direction perpendicular to the flow direction to obtain the corresponding one-dimensional curve, wherein the first region is either the C-line ROI region or the T-line ROI region.

[0140] The inflection point calculation module 1030, for example, performs operation S3 to divide the one-dimensional curve into two sub-curves with the highest point of the one-dimensional curve as the boundary, and calculates the inflection point of each sub-curve.

[0141] The generation and update module 1040, for example, performs operation S4 to generate two boundary lines based on the inflection points obtained in the first region, and uses the region between the two boundary lines as the new first region to obtain a new C-line ROI region and a new T-line ROI region.

[0142] The analysis module 1050, for example, performs operation S5 to analyze the concentration of the sample to be tested based on the gray values ​​of the new C-line ROI region and the new T-line ROI region.

[0143] The chromatography-quantitative rapid analysis system 1000 is used to perform the above... Figures 1-9 The illustrated embodiment presents a rapid chromatographic quantitative analysis method. For details not covered in this embodiment, please refer to the foregoing. Figures 1-9 The rapid quantitative chromatographic analysis method shown in the embodiments will not be described in detail here.

[0144] This invention also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements... Figures 1-9The rapid quantitative chromatographic analysis method shown in the embodiments will not be described in detail here.

[0145] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rapid quantitative chromatographic analysis method, characterized in that, include: The immunochromatographic image of the sample to be tested is converted into a grayscale image, and the C-line ROI region and T-line ROI region are searched in the grayscale image; The first region is compressed either as a whole or in sections along the direction perpendicular to the flow direction to obtain a corresponding one-dimensional curve, wherein the first region is either the C-line ROI region or the T-line ROI region. Using the highest point of the one-dimensional curve as the boundary, the one-dimensional curve is divided into two sub-curves, and the inflection point of each sub-curve is calculated. Based on the inflection points obtained in the first region, two boundary lines are generated, and the region between the two boundary lines is taken as the new first region, resulting in a new C-line ROI region and a new T-line ROI region. The concentration of the sample to be tested is analyzed based on the gray values ​​of the new C-line ROI region and the new T-line ROI region.

2. The rapid quantitative chromatographic analysis method as described in claim 1, characterized in that, When the first region is compressed as a whole: there is a one-dimensional curve corresponding to the first region; there are two inflection points obtained in the first region; the two generated boundary lines are straight lines that pass through the inflection points and are perpendicular to the direction of liquid flow.

3. The rapid quantitative chromatographic analysis method as described in claim 1, characterized in that, When the first region is compressed by region: there are N one-dimensional curves corresponding to the first region, N≥2; and there are 2N inflection points obtained in the first region. The step of generating two boundary lines based on the inflection points obtained in the first region includes: smoothly connecting the inflection points located on the same side of the highest point of the one-dimensional curve in sequence, and using the two curves obtained by the connection as two boundary lines.

4. The rapid quantitative chromatographic analysis method as described in claim 1 or 3, characterized in that, The first region is compressed by region, including: Vertically along the direction of the liquid flow, the first region is divided into groups every k columns of pixels to divide the first region into N sub-regions, where 1≤k≤7. The sub-regions are compressed as a whole along the direction perpendicular to the flow direction to obtain a one-dimensional curve corresponding to each sub-region.

5. The rapid quantitative chromatographic analysis method as described in claim 1, characterized in that, The step of searching for the C-line ROI region and the T-line ROI region in the grayscale image includes: The grayscale image is sequentially compressed in the direction perpendicular to the flow direction and then subjected to moving mean filtering to obtain the corresponding one-dimensional smooth curve. Search for the peak point of the one-dimensional smooth curve, where the value of the peak point satisfies: S peak >0.2*max(S)+0.8*min(S) Among them, S peak Where S is the value of the peak point, and S is the one-dimensional smooth curve; Find the peak endpoints along both sides of the one-dimensional smooth curve from the peak point, and determine the C-line ROI region and the T-line ROI region based on the peak endpoints and the peak point.

6. The rapid quantitative chromatographic analysis method as described in claim 1, characterized in that, The step of analyzing the concentration of the sample to be tested based on the gray values ​​of the new C-line ROI region and the new T-line ROI region includes: Calculate the C-line signal value based on the grayscale values ​​of the new C-line ROI region and the C-line background region; Calculate the T-line signal value based on the grayscale values ​​of the new T-line ROI region and the T-line background region; The concentration of the sample to be tested is calculated based on the ratio between the T-line signal value and the C-line signal value.

7. The rapid quantitative chromatographic analysis method as described in claim 6, characterized in that, The C-line signal value and the T-line signal value are respectively: C=|g b -g0| T=|g′ b -g′0| Where C is the C-line signal value, T is the T-line signal value, and g b g is the grayscale value of the new C-line ROI region, g0 is the grayscale value of the C-line background region, g ′ b For the grayscale value of the new T-line ROI region, g ′ 0 represents the grayscale value of the T-line background area.

8. The rapid quantitative chromatographic analysis method as described in claim 6, characterized in that, The C-line signal value and the T-line signal value are respectively: C=|lg(g b )-lg(g0)| T=|lg(g ′ b )-lg(g ′ 0)| Where C is the C-line signal value, T is the T-line signal value, and g b g is the grayscale value of the new C-line ROI region, g0 is the grayscale value of the C-line background region, g ′ b For the grayscale value of the new T-line ROI region, g ′ 0 represents the grayscale value of the T-line background area.

9. A rapid chromatographic quantitative analysis system, characterized in that, include: The conversion and search module is used to convert the immunochromatographic image of the sample to be tested into a grayscale image, and search for the C-line ROI region and T-line ROI region in the grayscale image; The compression module is used to compress the first region as a whole or in sections along the direction perpendicular to the flow direction to obtain the corresponding one-dimensional curve, wherein the first region is either the C-line ROI region or the T-line ROI region. The inflection point calculation module is used to divide the one-dimensional curve into two sub-curves with the highest point of the one-dimensional curve as the boundary, and to calculate the inflection point of each sub-curve. The generation and update module is used to generate two boundary lines based on the inflection points obtained in the first region, and to take the region between the two boundary lines as the new first region to obtain a new C-line ROI region and a new T-line ROI region. The analysis module is used to analyze the concentration of the sample to be tested based on the gray values ​​of the new C-line ROI region and the new T-line ROI region.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the rapid quantitative chromatographic analysis method as described in any one of claims 1-8.

Images